Improvement of the ridge estimation in regression linear model

Author

Peng, Deng ; Chun-sheng, Lin ; Tian-Hui, Fu

Author_Institution

Dept. of Weaponry Eng., Univ. of Naval Eng., Wuhan, China

Volume

1

fYear

2010

fDate

16-18 April 2010

Abstract

Under the morbid Gauss-Markov model, Ridge estimation(RE) is very effective in data processing. But the determination of ridge parameters is an issue. Ridge estimation shared the difficulty of the determination of ridge parameters. This paper brings about a method to solve the above problem, which is called generalized partial ridge estimation(GPRE). With it we can obtain the optimal solution to the GPRE directly and the ridge parameters are not to be calculated all. The method proposed in this paper provides a more effective technical to the biased estimation than ridge estimation. The mean square error of GPRE is smaller than RE and the mean square residual is larger.

Keywords

Gaussian processes; eigenvalues and eigenfunctions; matrix algebra; parameter estimation; regression analysis; GPRE; data processing; generalized partial ridge estimation; mean square error; mean square residual; morbid Gauss-markov model; regression linear model; Covariance matrix; Data engineering; Data processing; Eigenvalues and eigenfunctions; Gaussian processes; Least squares approximation; Mean square error methods; Parameter estimation; Vectors; Weapons; aeromagnetic; generalized partial ridge estimate; morbid equation; ridge parameter engeening;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Engineering and Technology (ICCET), 2010 2nd International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-6347-3

Type

conf

DOI

10.1109/ICCET.2010.5486182

Filename

5486182